BACKGROUND: Patients can experience recurrence following curative-intent resection of non-functional pancreatic neuroendocrine tumors (NF-pNETs). We sought to develop a nomogram to risk stratify patients relative to recurrence following resection of NF-pNETs.METHODS: Patients who underwent curative-intent resection for NF-pNETs between 1997 and 2016 were identified from a multi-institutional database. The impact of clinicopathologic factors, including tumor burden score (TBS) (TBS2=(maximum tumor diameter)2+(number of tumors)2), was assessed relative to recurrence-free survival (RFS), and a nomogram was developed and internally validated.RESULTS: With a median follow-up of 31.0months (IQR 11.3-56.6months), 66 (15.8%) out of 416 patients in the cohort experienced tumor recurrence. Overall, 3-, 5-, and 10-year RFS following curative-intent resection was 83.2%, 74.0%, and 44.7%, respectively. Several factors were associated with risk of recurrence including tumor grade (referent G1: G2, HR 4.07, 95% CI 2.29-7.26, p<0.001; G3, HR 10.83, 95% CI 3.72-31.53, p<0.001), lymph node metastasis (LNM) (HR 4.71, 95% CI 2.69-8.26, p<0.001), as well as TBS (referent low: medium, HR 4.36, 95% CI 2.06-9.24, p<0.001; high, HR 6.04, 95% CI 2.96-12.31, p<0.001). A weighted nomogram including tumor grade (G1 0, G2 54.19, G3 100), LNM (N0 0, N1 42.06), and TBS (low 0, medium 44.07, high 56.48) was developed. The discriminatory power of the nomogram was very good with a C-index of 0.75 (95% CI, 0.66-0.79) in the training cohort and 0.71 (95% CI, 0.65-0.75) in the validation cohort. In addition, the nomogram performed better than the current 8th edition of AJCC TNM staging system, which had a C-index of 0.67 (95% CI, 0.60-0.73).CONCLUSIONS: A nomogram that incorporated tumor grade, LNM, and TBS was established that had good discrimination and calibration. The nomogram may be an effective tool to stratify patients relative to recurrence risk following resection of NF-pNETs.
View details for DOI 10.1007/s00268-021-06020-8
View details for PubMedID 33768309